CN112741449A

CN112741449A - Automatic adjusting mattress

Info

Publication number: CN112741449A
Application number: CN201911048735.0A
Authority: CN
Inventors: 刘众; 王新安; 孙成俊; 卢坤涛; 何春舅
Original assignee: Peking University Shenzhen Graduate School
Current assignee: Peking University Shenzhen Graduate School
Priority date: 2019-10-31
Filing date: 2019-10-31
Publication date: 2021-05-04

Abstract

The invention relates to the technical field of intelligent adjusting beds, and particularly provides a sensor and an automatic adjusting mattress. This sensor includes the sensing structure, and this sensing structure includes the sensing layer, and this sensing layer top-down includes in proper order: the pressure sensing layer comprises a pressure film sensor, and a convex-concave structure or a hard material is arranged on one side above and/or one side below the sensing structure. The utility model provides a mattress sensitivity is high, uses the aspect, has really realized intelligent regulation.

Description

Automatic adjusting mattress

Technical Field

The invention relates to the technical field of intelligent adjusting beds, in particular to an automatic adjusting mattress and an intelligent bed comprising the same.

Background

In different sleeping positions of people, according to human engineering mechanics, the support needed by each part of the body is different, and the body pressure of each area is different, so that from the perspective of keeping health and balance, when lying down and lying on the side, the normal physiological bending of the spine needs to be kept. Conventional mattresses provide only simple support and are not intelligently adjustable.

Proper adjustment of the support of certain areas of the human body is required in view of the physiological curvature of the spine, the proper distribution of body pressure, and the smoothness of muscles and blood supply. Under this kind of condition, the mattress need satisfy simultaneously with low costsly, does benefit to the requirement of volume production and popularization, will light in weight simultaneously, does benefit to the transportation, but also needs to remain the originally comfort level of mattress. The existing scheme is that a multi-area height adjustment is achieved on a mattress in a mode of adopting a motor and a mechanical structure, meanwhile, a lattice sensing module is matched with the mattress, the requirements can be basically met theoretically, but the challenges are weight and cost, and finally the whole mattress is thicker in order to ensure comfort.

From a technical point of view, there is a need for sensing and real-time adjustment of the pressure at various parts of the user's body. But no intelligent adjusting mattress capable of well realizing the function exists in the products in the current market.

Therefore, in the current market, there is a great demand for a mattress that can be automatically adjusted according to the sleeping posture.

Disclosure of Invention

In order to solve the problems, the invention provides an intelligent mattress capable of being automatically adjusted, which aims to solve the problem that the existing mattress can not be adjusted in real time according to the sleeping body pressure, and simultaneously considers the cost, experience feeling, weight, industrialization realization difficulty and the like.

The embodiment of the invention is realized in such a way that the sensor for the mattress comprises a sensing structure, wherein the sensing structure comprises a sensing layer, and the sensing layer sequentially comprises from top to bottom: the pressure sensing layer comprises a lattice pressure sensor film, and a convex-concave structure or a hard material is arranged on one side above and/or one side below the sensing structure.

Another object of the embodiments of the present invention is to provide a mattress, which includes the sensor provided by the present invention, and further includes an air bag, wherein a comfort layer is disposed above the sensor, and a buffer layer is disposed below the sensor.

Another object of an embodiment of the present invention is to provide an air bag adjusting mattress, which includes a plurality of independent air bags, an air pump air valve, and a separate air tube, wherein the plurality of independent air bags are connected to the air pump of the air pump through the separate air tube, the separate air tube is provided with the air valve for controlling air flow, and a caliber of a connection portion of the separate air tube and the plurality of independent air bags is larger than a caliber of a connection portion of the separate air tube and the air pump.

The embodiment of the invention also provides an intelligent adjusting bed which comprises the air bag adjusting mattress.

The technical scheme of this application adopts the combination of gasbag and flexible sensing structure, and both correspond the function of regulation and sensing respectively. The air pump and the flexible sensor are not hard in texture, so that the requirement of comfort level is easily met.

Drawings

FIG. 1 is a mattress construction provided by one embodiment of the present invention;

FIG. 2 is a schematic diagram of the arrangement of the sensing layers in the sensor according to one embodiment of the present invention;

FIG. 3 is a schematic structural diagram of a sensing structure provided in accordance with an embodiment of the present invention;

FIG. 4 is a schematic structural diagram of a sensor structure with a cut buffer layer according to another embodiment of the present invention;

FIG. 5 is a schematic diagram of a mattress according to one embodiment of the present invention;

FIG. 6 is a schematic view of the structure of the divided air tube of FIG. 5;

FIG. 7 is a schematic view showing the structure of the air pump with a sound insulating device in FIG. 5;

FIG. 8 is a schematic circuit diagram of an embodiment of the present invention;

FIG. 9 is a flow chart of system control provided by an embodiment of the present invention;

FIG. 10 is a schematic diagram of a low pass filter structure used in one embodiment of the present invention;

FIG. 11 is a diagram of a low pass filter structure according to an embodiment of the present invention;

fig. 12 is a circuit configuration diagram of a multistage instrument amplification filter according to an embodiment of the present invention;

FIG. 13 is a sensing identification flow method according to an embodiment of the present invention;

fig. 14 is a flow chart of a torso recognition method according to an embodiment of the present invention;

FIG. 15 shows the steps of determining the sleeping posture according to the embodiment of the present invention;

FIG. 16 shows the specific steps for determining the symmetry of the torso region (SYM) according to an embodiment of the present invention;

fig. 17 is a schematic diagram of an adjustment algorithm of the intelligent adjustment bed according to an embodiment of the present invention.

Detailed Description

In order to make the technical problems, technical solutions and advantageous effects to be solved by the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The embodiment of the invention provides a sensor for an intelligent bed, which can be a piezoresistive sensor, a pressure-volume sensor or a strain gauge pressure sensor according to the type of materials.

Specifically, this sensor includes sensing structure, and this sensing structure includes the sensing layer, and this sensing layer top-down includes in proper order: the pressure sensing layer comprises a dot matrix thin film pressure sensor, and a convex-concave structure or a hard material is arranged on one side above and/or one side below the sensing structure.

In the embodiment of the invention, the mattress can be divided into the following parts according to the relation between the sensing module and the adjusting module: sensing and regulating independent structures (see fig. 1). Fig. 1 shows an embodiment of a mattress with independent sensing and adjustment structures, which comprises, from top to bottom, as shown in fig. 1: comfort layer, sensing layer, buffer layer and gasbag layer, wherein be provided with the buffer layer in the middle of sensing layer and gasbag layer (adjusting the structure), both are independent in space, each other do not influence. The air bags are inflated and deflated to adjust the height of the corresponding part of the mattress. The sensing layer mainly comprises: a flexible upper signal conducting layer, a flexible lower signal conducting layer, and an intermediate pressure sensing layer.

The sensing layer used in the embodiments of the present invention may be configured as a lattice thin film pressure sensor. For a dot matrix sensor, also called a planar dot matrix sensor, the sensor is a whole laminated flexible film, and inside the film is a matrix arrangement of distributed sensing units, as shown in fig. 2, which shows an arrangement of the sensing units, wherein the sensing units are arranged in an array, each row has a row of scanning lines, and each column has a signal output line. This arrangement allows the acquisition of a matrix-like pressure value of the sensor membrane.

FIG. 3 shows a diagram of a sensor layer profile in one embodiment of the invention, the sensor including a flexible top signal layer and a flexible bottom signal layer with a core sensor layer in between. The flexible top signal layer is also referred to as a flexible top routing layer and the flexible bottom signal layer is also referred to as a flexible bottom routing layer. The upper part of the sensing module is a comfortable layer, and the lower part of the sensing module is a buffer layer used for isolating the sensing layer and the adjusting layer. The independent sensing structure has the problems that the pressure borne by the surface of the mattress is transversely distributed to the periphery through the medium layer (comfort layer), so that the independence of sensing data is caused, and the pressure borne by the sensing layer has certain deviation with the body pressure of an actual sleeper, so that the detection result is inaccurate. In order to reduce such deviations, the invention proposes a cut-and-separate approach.

The cutting and separating method specifically means that the upper comfortable layer is cut by a certain thickness in the column direction of the sensing units (the cutting marks can penetrate through the comfortable layer and can not penetrate through the comfortable layer), so that the sensing units have better independence and are prevented from interfering with each other, and the specific structure is shown in fig. 4. The cutting lines of the comfort layer in fig. 4 are parallel to each other and to the arrangement direction of the cell sensors.

The mattress provided by the embodiment of the invention further comprises an adjusting device, the sensor detects the pressure condition borne by the mattress and then transmits information to the adjusting device, and the adjusting device performs adaptive adjustment on the surface height of the mattress so as to enable a user to obtain the most comfortable sleeping posture. The adjusting device used in the embodiment of the invention can be an air bag adjusting device, and can also be other adjusting devices.

The embodiment of the invention also provides an air bag type mattress adjusting structure which is as follows.

Example 1:

the mattress provided by the embodiment adopts an air bag adjusting mode, as shown in fig. 5, wherein the bottom of the mattress is formed by arranging a row of independent air bags, and each independent air bag is connected with an air pipe, so that the pressure and the lifting can be conveniently adjusted in a smaller area. The air bag and the air pipe are both connected to an air pump, and the middle of the air bag is controlled by an air valve. The operation of charging and discharging each air bag is controlled by the air valve and the air pump, thereby adjusting the hardness of each air bag.

On the other hand, in the embodiment of the invention, the separate air pipe is an air pipe with a variable caliber, as shown in fig. 6, the caliber of the air pipe at the independent air bag is larger, and the caliber of the air pipe at the air pump is smaller, so that the air speed of the air bag can be reduced during inflation and deflation under the condition that the air pump is under the same pressure, and the sound of the air flow in the air bag can be reduced.

Preferably, the air pump is disposed in the soundproof case and the buffer layer for damping vibration of the air pump and reducing sound of the air pump, as shown in fig. 7.

The improvement points of the embodiment of the invention are as follows: setting to reduce airflow sound: the caliber of the air pipe at the position of the separated air bag is larger, and the caliber of the air pipe of the air pump is smaller, so that the sound of air flow of the air pipe can be reduced.

Circuit structure

The mattress in the embodiment of the invention also comprises a circuit module structure which mainly comprises a sensing array scanning signal generating unit, a sensing signal amplifying and filtering unit, an AD unit, a Main Control Unit (MCU) and an air bag control unit, as shown in figure 8.

The circuit module adopted by the embodiment of the invention mainly comprises the following modules:

1. a filtering and amplifying unit: the method mainly comprises the steps of carrying out low-pass filtering and amplification on small signals output by a sensor array;

2. a signal scanning unit: forming a line scanning signal of the sensor;

an AD unit: AD processing is carried out on the filtered and amplified signals to convert the signals into digital signals;

an MCU unit: preprocessing the AD-processed sensing signals, and then performing a sensing recognition algorithm and an adjustment algorithm;

5. an airbag control unit: mainly a driving module for controlling an inflator of an air bag.

The system control flow of the embodiment of the present invention is shown in fig. 9.

Systematic key technical point of the invention

1. Signal noise processing

When a person lies still, the pressure is a static direct current signal, and the change of the pressure can be considered as the change of the direct current working point of the circuit. However, due to environmental factors such as power frequency interference or environmental micro-vibration, and due to circuit devices and power supplies, noise may be introduced at the output end of the sensing signal, and in severe cases, the noise of the signal may even drown out some effective signals with small amplitude. On the premise that the sensitivity of the sensor is constant, it is necessary to reduce the output noise of the signal. Therefore, the signal noise is reduced through circuit noise reduction and algorithm noise reduction, and the signal to noise ratio is improved.

(1) Circuit noise reduction: using low-pass filters

Since the pressure is a static signal and the frequency can be considered to be extremely low, and the environmental and circuit noises are white noises, that is, the noises are distributed in each frequency band, when the circuit is designed, the low-pass filtering is added at the sensing signal end, so that the environmental noises are reduced to a great extent.

As shown in fig. 10, a simple low-pass filter is provided, when the impedance of the sensor changes due to deformation, the output dc level changes, and the low-pass filtering structure composed of the resistor R and the capacitor C can realize low-pass filtering of the first order with the cut-off frequency f₀:

Fig. 11 shows a low-pass filter circuit diagram of another sensing structure. Wherein R1, R2, R3 and R4 form a group of full-bridge structures, the resistance values of R1 and R2 are opposite in change direction, the resistance values of R3 and R4 are opposite in change direction, and the resistance values of R1 and R3 are opposite in change direction. When a pressure is applied to the sensor, a voltage difference is induced across the operational amplifier +, -, and RC constitutes a low-pass feedback loop with a cut-off frequency f₀Comprises the following steps:

fig. 12 shows a structure diagram of a multi-stage instrument amplifying filter circuit with multi-stage low-pass and high-precision, wherein R1-R2-R3-R4-R, R1, R2, R3 and R4 form a group of sensors in a full-bridge structure, the resistance values of R1 and R2 are opposite in changing direction, the resistance values of R3 and R4 are opposite in changing direction, and the resistance values of R1 and R3 are opposite in changing direction. Wherein C1 and the output resistance of the sensor will form a first order low pass filter with a cut-off frequency f₁:

R5 and C2, R6, and C3 form a second-order low-pass filter, where R5 is R6, and C2 is C3. Cutoff frequency f2 of second-order low-pass:

at the output of the final stage of the amplifier, R13 and C4 are subjected to passive low-pass RC filtering of the third order, with cutoff frequency:

the third order can be arbitrarily selected to be a first order or a plurality of orders for filtering.

(2) And (3) performing algorithm noise reduction:

and setting a specific threshold value according to the characteristics of the sensing structure and the actually measured sensing value database, and processing noise.

The main algorithm noise reduction is to process digital signals after AD, and the noise reduction method mainly comprises the following steps: signal digital filter, mean smoothing filter and image denoising method.

The signal digital filter filters power frequency interference by adding a multi-order FIR or IIR low-pass filter, wherein the cut-off frequency can be 1HZ to 100Hz, and a 50Hz power frequency trap.

The mean filtering is to remove the glitch noise in some signals, i.e.

(3) Denoising an image:

and (3) regarding the sensing data as picture data, wherein the data of the sensing dot matrix corresponds to the value of each pixel point on the picture, and denoising by adopting a related denoising algorithm of the image, wherein the denoising algorithm comprises morphological open operation or morphological closed operation and the like. Namely, the image formed by the sensing data is subjected to binarization processing, and then expansion-first and corrosion-second or corrosion-first and expansion-second methods are carried out, so that the data is subjected to de-noising processing.

2. Recognition algorithm

The embodiment of the invention also provides a sensing identification method, which comprises the following steps: the number of people, whether the people are overweight or not, the body movement and the sleeping posture identifying and adjusting algorithm are specifically shown in figure 13.

1. Sensing matrix data acquisition

The pressure of each sensing area is converted into voltage by the sensing array, and then the voltage is converted into a data signal by the ADC, so that a pressure matrix S [ r ] [ c ] of the whole bed is obtained, wherein r and c are the row number and the column number of the array respectively.

2. Presence or absence of human recognition

Each pressure value of the pressure matrix is judged, and if all the pressure values are smaller than the unmanned judgment threshold Sth1, the pressure matrix is judged to be unmanned.

3. Overweight identification

Summing all the elements of the matrix, Sum (S [ i ] [ j ]), and if the threshold Sowth1 is exceeded, determining that it is overweight; or if a certain element S [ i ] [ j ] in the matrix exceeds the second threshold Sowth2, it is also determined to be overweight.

4. People number identification

And solving a transverse gradient, gradSx, of the matrix S, taking an extreme value of gradSx [ j ] of each row, and judging the number of people according to the number of the extreme values.

5. Recognition of whether to lie normally

Firstly, the pressure of the block is converted into an image, firstly, the edge of the image is identified, and then, the pressure points are clustered, so that the pressure block is obtained by segmentation. Then, trunk recognition is performed on the block, and one method of trunk recognition is to find the width of each row, then use the number of rows with similar widths as trunk parts, then find the length, width and aspect ratio of the trunk parts, and if the length, width and aspect ratio are within a certain range, determine the sleeping posture of normal lying, as shown in fig. 14.

6. Sleeping posture recognition algorithm based on SVM machine learning

The acquired matrix pressures are classified by a Support Vector Machine (SVM), and the sleeping postures are classified from 0 to 180 degrees according to the included angle between the normal vector (direction: from the back to the chest) of the plane of the trunk and the normal vector (direction: right side up) of the bed surface, and can be classified into a plurality of grades, such as 0 degree, 45 degrees, 90 degrees, 135 degrees and 180 degrees. Since the torso part of the human body is symmetrical left and right, it is not distinguished whether the left torso half part is the supporting shaft or the right torso half part is the supporting shaft, because, for example, 45 ° with the left torso half part as the supporting shaft and 135 ° with the right torso part as the supporting shaft are equivalent. Therefore, by default, the left half is taken as the support axis, and it can be found that 0 ° is supine, 180 ° is prone, and 90 ° is lying on the front side. Therefore, we can divide the sleeping postures into 5 types from the angle between the normal vector of the trunk and the normal vector of the bed surface.

The specific steps are shown in fig. 15, wherein:

s1: collecting a pressure matrix of a training sample in the 5 middle sleeping position;

s2: adopting a trunk part identification algorithm in the step (5) to identify the head and neck region, the trunk region, the hip region and the large and small leg region of the person;

s3: calculating the eigenvalues of the pressure matrix, the eigenvalues mainly comprising:

(1) the pressure specific gravities of the four body regions (head and neck region, torso region, buttocks and calf region),

(2) the gradient magnitude and gradient direction of the pressure pixel, e.g. the gradient of pixel (i, j):

Gi(i,j)＝F(i+1,j)-F(i-1,j)；

Gj(i,j)＝F(i,j+1)-F(i,j-1)；

G(i,j)＝sqrt(Gi(i,j)²+Gj(i,j)²)；

γ(i,j)＝arctan(Gj(i,j)/Gi(i,j))；

(3) trunk area symmetry SYM:

the calculation flow of SYM is shown in fig. 16:

(4) symmetry SYM of the hip region;

(5) variance of center of gravity of each row of leg regions.

S4, using the characteristic values of the sleeping postures obtained in the S3 as the input of the SVM classifier

S5: finding the optimal classification hyperplane (w X) + b ═ 0 of the classifier between every two sleeping postures, and constructing Lagrangian function

According to the duality of the Lagrangian function, the problem is converted into:

s.t.α_i≥0，i＝1，2，…，n

and finally, obtaining an optimal solution to obtain a decision function.

S6: a decision function between every two sleeping positions is obtained,

s7: and counting the votes, and recording the number of the votes as the finally classified sleeping postures.

7. The specific steps of the adjustment algorithm are shown in fig. 17:

s1: acquiring basic information of a user, comprising the following steps: sex, height, weight, BMI, length of torso, shoulder width, chest circumference, waist circumference, hip circumference, etc.

S2: and (3) establishing a muscle model and a vertebra model under a static state, namely, not considering the spatial position relation between the trunk and the bed, wherein the length direction of the trunk is consistent with the length direction of the bed, and the included angle between the normal phase of the trunk and the normal phase of the bed is 0 as a reference. The muscle model is the size of each part of the trunk, the spine model is the length of the spine, and the curvature of several key nodes (cervical, thoracic, lumbar and caudal).

S3: and identifying the positions of all areas of the user and the included angles between the body direction and the length direction of the bed through the sleeping posture.

S4: and (6) acquiring the included angle between the normal phase of the trunk and the normal phase of the bed surface through the sleeping posture identification in the step 6.

S5: by combining S2, S3 and S4, a space model of the human body trunk and the bed can be obtained, namely, the positions of the human trunk, including the positions of muscles and the spine under the reference frame of the bed, are determined.

S6: and calculating the adjusting height of each pixel dot matrix, wherein the adjusting height can balance the proper adjusting height under the muscle model and the adjusting height under the vertebra model to be used as the final adjusting height. The final aim is to obtain the best fit between the bed and the vertebra and the best force support for each part of the vertebra, and of course, to take the comfort of the muscles into account.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.

Claims

1. A sensor comprising a sensing structure, the sensing structure comprising a sensing layer, the sensing layer comprising, in order from top to bottom: the pressure sensing layer comprises a pressure film sensor, and a convex-concave structure or a hard material is arranged on one side above and/or one side below the sensing structure.

2. The sensor of claim 1, wherein the sensor comprises a sensing region and a non-sensing region, the sensing region having a greater modulus of elasticity than the non-sensing region.

3. Self-adjusting mattress comprising a sensor according to claim 1 or 2, further comprising an air bag, a comfort layer being arranged above the sensor and a buffer layer being arranged below, a cut of a certain thickness being made on the comfort layer, the cut either penetrating or not penetrating the comfort layer.

4. The self-adjusting mattress of claim 3, wherein the pressure sensing layer is a pressure film having pressure sensing points distributed in a lattice pattern within the sensor film.

5. An air bag adjusting mattress comprises a plurality of independent air bags, air valves and air pipes,

two adjacent independent air bags are communicated through air pipes at two ends, an air valve is arranged on each air pipe to control the flow of air flow, the independent air bags are integrally divided into three parts which respectively correspond to the head part, the spine part and the leg and foot parts, and the air flows in the independent air bags in the three parts can be communicated with one another.

6. The air bag conditioning mattress of claim 5, further comprising an air pump module connected to the air bag via an air valve.

7. An intelligent adjustment bed comprising an automatically adjusting mattress according to claim 3 or 4 or an air bag adjusting mattress according to claim 5 or 6.